Viewing profile — qaq
qaq
HN member- Joined
- Mon, Mar 16, 2015, 7:55 PM UTC
- HN karma
- 7,133
- Public activity
- 3,376 items
- HN profile
- View on Hacker News ↗
About qaq
No profile information was provided.
Recent public activity
-
comment
Comment #49223401
there was pretty large intelligence side that did attribution and investigative type work into orgs money movements etc. some of those people def came from non tech backgrounds.
-
comment
Comment #49223295
If one wants a conspiracy theory in USSR a First Deputy Chairman of the KGB offed himself when an investigation he was coordinating led to the daughter of General Secretary of the …
-
comment
Comment #49223192
Only one job ? Worked with a ton of ex NSA and CIA people in a large cybersec company.
-
comment
Comment #49191143
is this sarcasm ?
-
comment
Comment #49162101
AWS made 46B profit last year and will make prob close to 70B this yea so even 400B is very far from "world ending"
-
comment
Comment #48979441
well they traded short term gain for long term pain.
-
comment
Comment #48849385
I use both especially for checking each others work. Pretty happy with results
-
comment
Comment #48680284
API pricing is def not below cost
-
comment
Comment #48674229
If people are sure they can always short NVIDIA
-
comment
Comment #48674178
Most of Anthropic revenue looks to be companies paying for Claude Code at API prices ...
-
comment
Comment #48674149
hence the bit about us learning the actual state of things once they are a public company.
-
comment
Comment #48673761
A lot of factors there we will see how it plays out.
-
comment
Comment #48673505
We will see once they go public Dario did claim profit margin on inference is 40% if memory serves me right
-
comment
Comment #48673248
Yes 65B ARR that Anthropic has is clear indication there is no path to revenue.
-
comment
Comment #48673179
If people were not consuming their services they would not be buying inference hardware at this rate so it's pretty much on consumers.
-
comment
Comment #48641104
Fable was able to oneshot pretty big features. In write spec -> refine spec -> create todos -> implement todos workflow difference was far less pronounced vs codex or opus.
-
comment
Comment #48504286
Well given dual class shares in SpaceX Elon will get full control of Tesla for free aside from all other benefits ...
-
comment
Comment #48481763
it is crazy
-
comment
Comment #48480514
why did we switch to per minute? A modern quality enterprise SSD can do 35K +/- legit fsyncs per second.
-
comment
Comment #48467319
You don't need LLM for that. You make _all_ projects low-stakes by working on green field project using (insert buzzword soup of the day) and leaving for a new green field opportun…
-
comment
Comment #48451359
Basically xAI is pulling a Palantir. They try to reposition datacenter capacity lease revenue to have a multiplier of fast growing frontier lab.
-
comment
Comment #48451323
Where did you get 2-3B from? Colosus 2 GPU's alone were 18B Total cost including construction, power and water treatment facility might be close to 25-30B.
-
comment
Comment #48449667
Anthropic has made $330 billion in compute and chip commitments between Google, Amazon, and Microsoft, another $30 billion with CoreWeave and another $15 billion with SpaceX. To pa…
-
comment
Comment #48427514
None is using grok so they are renting out unused capacity
-
comment
Comment #48427505
They know allocation they have from TSMC for TPUs production they know allocation they have from Nvidia they see demand curve.